Aɣrab (yiwen n usentel)
Tixbula
Send ad testeqsiḍ, Ɣer uqbel isteqsiyen FAQ.
Iswi-nneɣ d akken amḍiq-a ad yili d win n talwit d amḍiq n leqdic zeddigen. Ttxil ɣretilugan-nneɣ mgal n wayen isaxsaren.
gillux
iḍelli
gillux
iḍelli
LeviHighway
iḍelli
EugeneGS
iḍelli
Ooneykcall
iḍelli
LeviHighway
iḍelli
frpzzd
iḍelli
sharptoothed
iḍelli
EugeneGS
iḍelli
Thanuir
iḍelli
We have a translation challenge for all of you!
http://tatoeba.org/eng/sentences/show/763114
Let's have it translated and translated for helping TrangTrang!!
Where can one download a program to use IRC?
Quite everything: http://pidgin.im/download/
Mozilla: https://addons.mozilla.org/it/f...don/chatzilla/
Microsoft: http://www.mirc.com/
Mac OS: http://colloquy.info/
:)
Trusted users & Tatoeba day
Since tomorrow is Tatoeba day, the trusted users voting session will be a bit special.
1) It will start tonight at 10PM, right before Tatoeba day begins.
2) It will carry on until the end of Tatoeba day.
It doesn't mean we will not sleep and debate about trusted users for 26 hours :P It just means there's a larger time window to discuss about who can become or not trusted user :)
Priority goes to people who've sent me a private message to express their wish to become trusted user. But this weekend, since I'm willing to dedicate more time to the issue, we will also try to go through the cases of people who have been nominated, but who haven't written to me that they want to become trusted users (and may not know that they have been nominated, or may not even know that there is a trusted user status).
Again, the list is in my profile (at the end):
http://tatoeba.org/eng/user/profile/TRANG
Sorry if the question is stupid, I am a freshman here.
I wanted to change the furigana in this sentence: http://tatoeba.org/eng/sentences/show/209006
since it should be 「しょうぼうふ」 rather than 「しょうぼうおっと」.
How can I accomplish that?
[not needed anymore- removed by CK]
Got it, thank you for the tip.
I have a question about restricting one's search.
I want to search for the word "重なる". In order to find all verb forms, I decided to search for "重な", however this finds a lot of entries with "貴重な". So I wanted to remove all these entries from my search, entering:"重な" -"貴重"
However, this led to no search results, although there are entries.
Can someone tell me the correct syntax? I know that it used to work like this.
[not needed anymore- removed by CK]
or under linux you can do this
grep 重な sentences.cvs | grep -v 重な
Yeah, but why can't I do it in the tatoeba search?
It used to work.
(重な) !(貴重) "重な"
This working for me. Only 重な, some results are 慎重.
If I use only
(重な) !(貴重)
it would search for "重な" "な", so lots of results with "な", but without 重 kanji.
I found the underlying issue:
When I search:
"重な" -"貴重" Language: Any
I get 0 results!
When I search:
"重な" -"貴重" Language: Japanese
I get 19 results!
I thought that setting the language to Any would also find the Japanese sentences!
hmmm strange, I will try to ask the guys of sphinx (the system we're using) (though i'm building my own one right now to replace it^^) I think I know why but i'm not sure
Vortarulo gave me an idea: why don't we use a script to turn
-- (two hyphens)
into
— (an em dash)?
Better yet, why don't we have special buttons along with the translation box for special characters like en and em dashes, smart quotes, etc?
answer: because I haven't found the time to code it yet :) But I do agree it will be usefull :)
Note to those in charge of organizing tags: "non-strandard orthography" is not exactly standard orthography.
:D
It’s redundant anyway, as means the same thing as the ‘non-standard spelling’ tag.
I got rid of the three uses of ‘non-strandard orthography’ tag, as it seems I was responsible for coining it. :o
Maybe there is the need for connections other than just "is translation of or other version":
For example this sentence http://tatoeba.org/deu/sentences/show/242949 about "students" leads to two different translations into german "Student" and "Schüler", meaning "university student" and "pupil" respectively.
So the two sentences are fairly similar, but also very different. I would like to discuss the possinility to connect these two as "is not the same as (although looking alike)".
There might also be thr need for "is the opposie of", "loose" connection or some other connections (that i cant make up for now). I guess tagging for connections was already suggested, but i doubt the visibility of such tags: we would need clear symbols like a stroke through "=" or an "~" or a "<=>" instead of the green arrow.
Maybe this would be too complicated, but i want it to be discussed. ;D
no for this the new version will introduce the possibility to have any kind of links, so far all "link" are "translation link", so it will be possible to have "not a translation of" link.
And may be make this links only visible on the sentence page (not when it's just inside a "block" (such as "random sentence" , search result etc) and in a "not translation of" part
Cool, seems like this time your concepts are quicker than my ideas! :D
This shoots into my mind that it would be cool to visualize the "whole" graph, enabling to jump from cluster to cluster along the connections or so. But i guess this would need a biiiiiig lot of coding unless we find some (open) tool to do this existing already. There must be other people who think about graph visualization out there, right?
Yesterday, I thought about a way to visualize the graph.
My idea is, to take one sentence in a certain language and all directly connected sentences.
These sentences could be displayed in a circle and between the single sentences all direct connections.
Around the circle are all sentences of the second level. I mean those, which are connected with sentences of the inner circle, but not connected with the initial one.
I've found a Flash mind map viewer (http://www.rubenswieringa.com/b...ctivemindmap/) on the blog of Ruben Swieringa ((http://www.rubenswieringa.com/b...tive-mindmap)) based on the SpringGraph component of Mark Sheperd (http://mark-shepherd.com/blog/s...x-component/).
This could be a basis for such a visualization.
Some cool stuff could be a way to easily draw new connections between the sentences or clean up duplication.
Yep but I don't plan to use flash, I'm quite a "web standard" nazi, It's a bit late where I am right now so I will try to find them tomorow, but I do remember to have some monthes ago a javascript library to handle this.
For duplication, thanks to the help of a friend who quite master this field, we've built a custom graph database server that will handle this automatically and in real time, so soon duplicate will be no more a problem :)
but what you said is basically what is produced by "dot", an open source software when you ask for output in SVG, but anyway tomorow I will try to find back this JS library.
A while ago, I started to make a graph visualization thing like you describe for Tatoeba. I kind of stopped though, figuring I'd wait for the API to come out, since it wasn't much use when I had to manually enter in sentence and link data.
I was using JIT: http://thejit.org/
Once the new version comes out, I'd be willing to finish this project. I also have a few other ideas that I'd like to code for Tatoeba. :-)
By the way, where is the code for the new version and database hosted? I remember looking at it once, but I can't find it now.
in fact it should not be a "lot" of coding, as there's already some xml like "language" for graph than can be after converted into SVG, the problem after is support from all browsers (does someone said IE-6/7/8 :p)
I would like to have a spell checker, but not for the obvious part when i am adding or editing a sentence, but for the sentences that are already there. There should be an easy way to enabled my browser built-in spellchecker on the sentences that are not inside a input field but just a plain link, right? What about checking multiple different languages at once, for example the esperanto and the german translation of an english semtence. I speak all these three, so i would want my browser to check them all with the proper languages. Also when typing i always have to switch to the proper spellchecker. Is there a way to add a html language tag to the imput field, so my browser will come up with the right dictionary? I'm going to do some research on this. I'm mainly thinking of firefox (or any other open browser) or an extension for it, but maybe there alreadynis such a thing.
I forgot: maybe we would want integrate a spell-checker to build up a dictionary of correct spellings by checking the red underlined words as okay. Maybe theres another open project for this purpose.
I'm on the way to release a project I've started last month to build a tatoeba-like dictionnary of word (well with some difference but I think you get the idea) so it would be possible (and that's why I'm making it anyway:p) to connect both, so tatoeba will be able to check if the words are present on the database of this project.
Cool, i guess it might grow very quickly. Quicker than an organized dictionary, where you have to fill in everything. It should grow naturally with the use of the words and those who are used often or need extra explanation can be commented on, right?
Willkommen bei Tatoeba, Bellinger.